The BridgeDb framework: standardized access to gene, protein and metabolite identifier mapping services

Abstract

Background: Many complementary solutions are available for the identifier mapping problem. This creates an opportunity for bioinformatics tool developers. Tools can be made to flexibly support multiple mapping services or mapping services could be combined to get broader coverage. This approach requires an interface layer between tools and mapping services.

Results: Here we present BridgeDb, a software framework for gene, protein and metabolite identifier mapping. This framework provides a standardized interface layer through which bioinformatics tools can be connected to different identifier mapping services. This approach makes it easier for tool developers to support identifier mapping. Mapping services can be combined or merged to support multi-omics experiments or to integrate custom microarray annotations. BridgeDb provides its own ready-to-go mapping services, both in webservice and local database forms. However, the framework is intended for customization and adaptation to any identifier mapping service. BridgeDb has already been integrated into several bioinformatics applications.

Conclusion: By uncoupling bioinformatics tools from mapping services, BridgeDb improves capability and flexibility of those tools. All described software is open source and available at http://www.bridgedb.org.

Figure 1

The BridgeDb architecture. BridgeDb provides a channel to connect multiple bioinformatics tools, such as Cytoscape or PathVisio, with online and offline identifier mapping services.

Figure 2

Transitive relationships between identifiers. In this diagram, two mapping sources (named A and B) are combined. Mapping source A defines relations between Ensembl and Entrez genes. Mapping source B is a custom microarray annotation that provides mappings between custom identifiers and Entrez. Through transitivity, the relation between the custom microarray and Ensembl can be inferred. The custom annotation file needs to define only a subset of relations.

Figure 3

Simplified UML Diagram of the BridgeDb java library. The BridgeDb.connect method serves as an entry point, instantiating one of the many IDMapper implementations. The most important method of IDMapper is mapID, which takes an Xref object as argument. An Xref is a combination of an identifier String and a DataSource object, the latter representing an online biological database.